Abstract The objective of this study was to evaluate the accuracy and precision of an electronic dosing apparatus (Gestal Multi – unit (GMU)) under commercial conditions and to identify potential sources of variation in supplement delivery. The study was conducted on a commercial sow farm in southwestern Minnesota equipped with the Gestal 3G feeding system (Jyga Technologies, St-Lambert-de-Lauzon, QC, Canada). Feeding stations (n = 88) were tested between January and June 2025 and provided 175 evaluations. Each observation represented the total daily feed dispensed for an individual sow across five feeding allowances that targeted a desired body condition. Individual drops were manually collected and weighed from both the feed delivery line and the doser line to determine the actual amount dispensed. For each feeding event, the feed and supplement were delivered in 3 to 6 drops, with each drop dispensing only a fraction of the total allotment. Delivery lines were calibrated before each testing period to ensure standardization. Data were analyzed using one-way ANOVA to evaluate the effect of feeding allowances on precision (deviation from the expected value) and accuracy (repeatability of delivery) and by simple linear regression to assess the relationship between total amount delivered and deviation from the programmed target of each allowance. The total feed delivered (feed delivery (Line 1) plus doser (Line 2)) was close to the expected values, with mean deviations ranging from –1.1% to –2.9%. Feeding allowances had no significant effect on percentage deviation for either total feed (P = 0.60) or supplement delivery (P = 0.16). Overall, the GMUs consistently underdelivered relative to target values (mean –10.5 g or –7%) which indicates a systematic shortfall. The magnitude of underdeliver increased as the quantity being delivered decreased. Sampling date significantly affected accuracy of delivery (P 0.02), which suggests that environmental factors such as temperature, humidity, or feed characteristics (e.g., particle size, density, or moisture content) influence target product delivery potentially by influencing consistency of material flow. The observed 7% shortfall in delivery of the supplement from the GMUs was predictable across feeding allowances. Depending on the intended use and management, this level of deviation may be inconsequential and simply accounted for in the settings of the operating software. The GMU system demonstrated high repeatability under commercial conditions. However, delivering small quantities ( 125 g) as a proportion of the total allowance from the GMU may require slight upward calibration of the programmed target to achieve the desired output. In practice, the GMU dispensed feed as several smaller drops, with an average of approximately 32 g per drop. Ongoing analysis will quantify precision and variability at the individual drop level.
Berrio et al. (Wed,) studied this question.